Re: [LAU] Whysynth and MIDI channels

From: Nick Copeland <nickycopeland@email-addr-hidden>
Date: Wed May 30 2007 - 22:02:33 EEST

Hi Fons,

It looks like preaching to the converted I admit.

>Because in all but the simplest cases that is not what you want.

Quite. However the person that wants the simplest case is the one that does
not want to spend time reading all the different readme files - this was
exactly the point of the submit to this list regarding the complexity of
just testing all these softsynths. If the default allows this connectivity
which would be very much in the mood of the original MIDI specification, as
opposed allowing no connections at all, then another user is catered for and
the design is in line with the MIDI specification.

>If you have 10 sources and 10 sinks,

Then you should also know how to break all the existing connections. Put it
this way, if you have worked out an easy way to setup all these connections
then you should have been able to fid a similarly easy way to disable any
default MIDI mixed connections that you did not want.

On the other hand if you don't know how to build all your ten connections
then you probably don't even know how to build one and you are in for a
world of hurt just getting a sound out of a softsynth. That is totally
counterproductive. Unless the default were to connect things together.....

>what is easier, setting up 10 connections
>or working out which 90 to disconnect ? Is forcing a user to do N-1 dis-
>connects each time he starts something new making his work easier ?

I am not talking about making this work easier, however it is not making it
that much harder for an experienced user either. This issue here is that
when you know how do to something it is easy, when you don't know how to it
is hard, and the default should be to reduce complexity.

>I want to be able to start an app and be sure it will not interfere
>with what is already running. So it should not auto-whatever.

Then as a capable user you should have started the whole connection toolkit
without the defaults and just built your connections. Listen up friend, if
you are in charge of a PA system and even trust that the default connection
setup is "none" then you are in a dangerous position - you should make sure
of it first by breaking connections before bring things online. You are
again making assumptions and these ones might end up blowing your speakers
for you. Oh, yeah, deja vu.

>Think about this. Many MIDI devices will echo their input on the output.

That is MIDI Thru and was intended to be kept that way. I don't know of many
MIDI devices that will echo In to Out as a default since the default for
this functionality is the Thru link - again, the capable user may configure
this if their device supports it, but then they should know about the
implications anyway. Interestingly enough, from your last submit to this
list you refered to this very function as a 'trick' that if used would break
things, now you are quoting the same MIDI mixing function (in this case
mixing In with whatever you play on this synth) as if it were a standard
feature which is kind of contradicting yourself, no?

Back to my original point, the MIDI specification stated that a system
should start up in OMNI mode unless otherwise specified such that any
keyboard could drive any sound module on power up since channel numbers
would not be considered. A capable user could change this configuration, but
that is their choice.

My proposal remains true to the mood of the original specification, here
suggesting that a MIDI seq interface should consider being OMNI as well and
allow things to be connected together by default.

Regards,

N.

_________________________________________________________________
FREE pop-up blocking with the new MSN Toolbar - get it now!
http://toolbar.msn.click-url.com/go/onm00200415ave/direct/01/

_______________________________________________
Linux-audio-user mailing list
Linux-audio-user@email-addr-hidden
http://lists.linuxaudio.org/mailman/listinfo.cgi/linux-audio-user
Received on Thu May 31 00:15:04 2007

This archive was generated by hypermail 2.1.8 : Thu May 31 2007 - 00:15:05 EEST